Individual Participant Data Meta-Analysis. Группа авторов
and safety of more intensive lowering of LDL cholesterol, eligibility was restricted to those randomised trials that aimed to recruit 1,000 or more participants with a treatment duration of at least two years.89 However, depending on the research question, it should be noted that the information size for a trial usually depends on more factors than the total participants (Section 2.6.3), and that statistical power calculations can be used in advance of IPD collection to more appropriately identify which trials provide the most information (Chapter 12), and thereby indicate which should be prioritised for data collection.
Similarly, if older trials are potentially eligible, but might be particularly difficult to track down or are less applicable, then it might be appropriate to exclude them. For example, older trials using long‐term alkylating agents were excluded from an IPD meta‐analysis of the effects of adjuvant chemotherapy for non‐small‐cell lung cancer,90 because previously they had been found to be harmful to patients,12 and therefore were no longer of relevance to the clinical question.
The central research team might also prioritise seeking IPD from the trials deemed to be of higher quality, based on a formal risk of bias assessment,91 in order to focus their efforts on trials that would provide the most reliable IPD meta‐analysis results (Chapter 9). At the project design stage, necessarily, these judgements would have to be based on information obtained from trial protocols, publications, reports or trial investigators, rather than the IPD itself (Section 4.6).
4.2.2 Developing the Protocol for an IPD Meta‐Analysis Project
As for all research studies, a protocol is an invaluable tool for working through the design and guiding the conduct of an IPD meta‐analysis project. It serves not only to describe the planned methods in advance of data collection, but is also a main vehicle for communicating the rationale for the project, and convincing potential collaborators of the value of participation – reassuring them that the project is sensible and methodologically sound, and that their IPD will be used appropriately. However, if funders wish to receive or sign off a protocol before releasing project funds, for example, this can pose practical difficulties. A way around this may be to provide funders with a project scope (Section 3.3) or draft version of the protocol, and explain that it will be finalised after collaborators have had the opportunity to comment. This then allows the central research team to modify plans according to the feedback received, and describe which data the included trials have actually collected and will make available.
Figure 4.2 PICOS example: objective and eligibility criteria for an IPD meta‐analysis of pre‐operative chemotherapy for non‐small‐cell lung cancer.88
Source: Based on NSCLC Meta-Analysis Collaborative Group. Preoperative chemotherapy for non-small cell lung cancer: a systematic review and meta-analysis of individual participant data. Lancet 2014;383:1561–71.
Key sections to include in the project protocol are shown in Box 4.1. Although the structure is broadly similar to that for a conventional systematic review and meta‐analysis of existing aggregate data, greater attention should be paid to explaining the rationale for seeking IPD, describing how data will be checked and used, and, in particular, detailing the analyses that will be performed. The latter is particularly important, because IPD offer greater analytical breadth, flexibility and complexity, thereby introducing more potential for inappropriate analyses, and also more opportunity for ‘data dredging’, data manipulation or the selection of analyses that will generate particular findings, or support prior beliefs. Developing a detailed and publicly accessible protocol prior to receiving IPD helps safeguard against these practices, and at the same time counters any unfounded suspicion or allegations of such manipulation. At a minimum, this would include the main and additional outcomes, and their definitions; methods of individual trial analysis (if relevant) and meta‐analysis, including those for exploring potential effect modifiers at the trial or participant level; and methods for quantifying and accounting for heterogeneity.
Box 4.1 Key sections to include in an IPD meta‐analysis protocol
The rationale, including reference to any prior motivating systematic review or meta‐analysis, and why IPD are needed
A description of the research objectives
The inclusion and exclusion criteria relating to participants, interventions, comparisons, outcomes and study design types (i.e. PICOS), including whether these will be applied at the trial level or participant level, and any other factors such as the qualifying dates for trial completion
A summary of the methods used for trial identification, including bibliographic search strategies and any additional means of identifying eligible trials, such as communication with trial investigators, along with information about the screening process
A description of how IPD received will be checked and verified, and how trial quality and the risk of bias will be assessed based on the IPD
Details of the participant‐level variables to be collected
A list of main and additional outcomes and their definitions
A summary of the statistical approaches and models to be used, both for analysing trials separately (if applicable) and for IPD meta‐analysis, including details of methods for: accounting for clustering of participants within trials, quantifying and exploring heterogeneity, modelling participant‐level variables (and any interactions), and any sensitivity analyses. These should be specified in sufficient detail in the protocol or in a separate statistical analysis plan (SAP), in order to permit scrutiny and, if necessary, replication (through access to the same IPD)
An initial power calculation or sample size justification, conditional on assuming IPD are available from particular trials
An approximate timetable for the project
Details of members of the central research team, the project advisory group or steering committee, patient and public involvement representatives and funders
A dissemination plan
A provisional table of eligible studies
Appendices of, for example, search strategies for individual databases and the data dictionary
Source: Jayne Tierney and Lesley Stewart.
As for other clinical studies, an additional statistical analysis plan (SAP) may be required, in which case, the protocol should focus on the general methods and a summary of the planned analyses, whilst the SAP would provide more detailed explanation of statistical methods and modelling techniques (see guidance in Part 2). The anticipated scale and complexity of the project will help guide whether the protocol is sufficient, or whether an additional SAP is needed.
Although the protocol and SAP should pre‐specify the analyses in some detail, this does not preclude analyses that become necessary to further explore, explain, or add to the main findings. The protocol should make it clear that such analyses may be needed, and